ClonEval: An Open Voice Cloning Benchmark
Christop, Iwona, Kuczyński, Tomasz, Kubis, Marek
–arXiv.org Artificial Intelligence
ABSTRACT We present a new benchmark for voice cloning text-to-speech models. The benchmark consists of an evaluation protocol, an open-source library for assessing the performance of voice cloning models, and an accompanying leaderboard. The paper discusses design considerations and presents a detailed description of the evaluation procedure. The usage of the software library is explained, along with the organization of the leaderboard. The evaluation results of selected open-source models are reported.
arXiv.org Artificial Intelligence
Sep-18-2025
- Country:
- Europe > Poland
- Greater Poland Province > Poznań (0.40)
- North America > Canada
- Europe > Poland
- Genre:
- Research Report (0.70)
- Industry:
- Information Technology > Security & Privacy (0.87)
- Technology:
- Information Technology > Artificial Intelligence
- Machine Learning > Neural Networks (0.97)
- Natural Language (1.00)
- Speech (1.00)
- Vision (1.00)
- Information Technology > Artificial Intelligence